Decentralized Federated Learning (DFL) offers a scalable paradigm for collaborative intelligence at the edge, yet its practical efficacy is severely constrained by system heterogeneity. Traditional synchronous protocols enforce rigid, lockstep aggregation barriers, where the training velocity of the entire collective is strictly dictated by the slowest straggler node, inevitably leading to significant idle time and resource underutilization. While asynchronous strategies mitigate latency, they often introduce complex pathologies, such as unbounded staleness and systemic unfairness, because high-performance nodes disproportionately bias the global model toward local data distributions, thereby marginalizing slower contributors. To rigorously reconcile these conflicting trade-offs, this work presents CAFF , a novel asynchronous communication framework for DFL that decouples local optimization from global synchronization via a topology-aware, event-driven protocol. By implementing a topology-aware cache with a strict per-neighbor replacement policy, the mechanism limits per-peer dominance by enforcing a one-slot-per-neighbor cache and exclusive replacement, preventing any peer from contributing multiple updates within a single aggregation event. Furthermore, a configurable staleness filter and a dynamic aggregation threshold ensure robust convergence stability across diverse federation topologies. Extensive empirical evaluations using MNIST, FashionMNIST, CIFAR-10, and SVHN, conducted on a high-fidelity, virtualized testbed across fully connected, star, and ring topologies, demonstrate that CAFF significantly outperforms synchronous baselines. Specifically, in dense network configurations, the framework reduces wall-clock training time by up to 39% and network traffic by up to 75%, while maintaining competitive predictive fidelity with controlled accuracy degradation. These results position CAFF as a robust and scalable efficiency-oriented solution for heterogeneous peer-to-peer learning environments.

Decentralized Federated Learning (DFL) enables nodes to collaboratively train models without a central server, introducing new vulnerabilities since each node independently selects peers for model aggregation. Malicious nodes may exploit this autonomy by sending corrupted models (model poisoning), delaying model submissions (delay attack), or flooding the network with excessive messages, negatively affecting system performance. Existing solutions often rely on rigid configurations or additional infrastructure, such as blockchains, which can incur computational overhead, introduce scalability issues, or limit adaptability. To overcome these limitations, this paper proposes RepuNet, a decentralized reputation system that categorizes threats in DFL and dynamically evaluates node behavior using metrics like model similarity, parameter changes, message latency, and communication volume. Node influence in model aggregation is adjusted based on each node’s reputation score. RepuNet was integrated into the Nebula platform and experimentally evaluated with MNIST and CIFAR-10 datasets under non-IID distributions, using federations of up to 25 nodes in both fully connected and random topologies. The evaluation considers different attack intensities, frequencies, and activation intervals, and includes comparisons with Byzantine-resilient aggregation mechanisms (Krum and Trimmed Mean), stronger structured poisoning strategies (Signed Neuron Remapping and GLL Neuro Inversion), as well as an ablation study of the exclusion threshold and a communication overhead analysis. Results demonstrate that RepuNet effectively detects and mitigates malicious behavior, achieving F1 scores above 95% on MNIST and approximately 76% on CIFAR-10. These outcomes highlight the adaptability, robustness, and practical potential of RepuNet for mitigating threats in decentralized environments.

Federated Learning (FL) offers a paradigm for collaborative AI that mitigates raw data exposure, yet the statistical heterogeneity of client data severely constrains its practical application. This non-independent and identically distributed (non-IID) setting induces client drift, leading to unstable optimization and degraded generalization, particularly for under-represented classes. Existing solutions present a difficult trade-off: iterative, regularization-based methods suffer from high communication overhead and a centralized bottleneck, while knowledge-distillation-based approaches rely on impractical artifacts, such as shared public datasets. This work introduces FedEnD , a novel framework that addresses the previous challenge through an efficient, fully decentralized architecture. FedEnD employs a two-stage protocol that decouples local specialist training from a collaborative fusion stage. Following a communication-free training phase, clients perform a one-shot peer-to-peer broadcast that shares (optionally privatized) specialist parameters and lightweight class-count statistics. Each client then uses these statistics to construct a class-distribution-weighted teacher ensemble from the received specialists. Crucially, this ensemble’s knowledge is distilled into a robust global model on each client, using only their local data as unlabeled inputs, obviating the need for a central server or auxiliary data. Extensive experiments on MNIST, FashionMNIST, SVHN, and CIFAR-10 demonstrate that FedEnD outperforms baselines, surpassing robust methods such as SCAFFOLD by +5.7% on complex datasets in pathologically skewed settings. This superior accuracy is achieved while reducing communication bandwidth by 68.6% compared to standard iterative averaging, and by up to 84% compared to gradient-correction methods like SCAFFOLD, highlighting a favorable trade-off between accuracy and communication bandwidth in decentralized learning under severe non-IID partitions.

Context: Decentralized Federated Learning (DFL) is an emerging paradigm that enables collaborative model training without centralized data and model aggregation, enhancing privacy and resilience. However, its sustainability remains underexplored, as energy consumption and carbon emissions vary across different system configurations. Understanding the environmental impact of DFL is crucial for optimizing its design and deployment. Objective: This work aims to develop a comprehensive and operational framework for assessing the sustainability of DFL systems. To address it, this work provides a systematic method for quantifying energy consumption and carbon emissions, offering insights into improving the sustainability of DFL. Methods: This work proposes GreenDFL , a fully implementable framework that has been integrated into a real-world DFL platform. GreenDFL systematically analyzes the impact of various factors, including hardware accelerators, model architecture, communication medium, data distribution, network topology, and federation size, on the sustainability of DFL systems. Besides, a sustainability-aware aggregation algorithm ( GreenDFL-SA ) and a node selection algorithm ( GreenDFL-SN ) are developed to optimize energy efficiency and reduce carbon emissions in DFL training. Results: Empirical experiments are conducted on multiple datasets, measuring energy consumption and carbon emissions at different phases of the DFL lifecycle. Results indicate that local training dominates energy consumption and carbon emissions, while communication has a relatively minor impact. Optimizing model complexity, using GPUs instead of CPUs, and strategically selecting participating nodes significantly improve sustainability. Additionally, using wired communication, particularly optical fiber, effectively reduces energy consumption during the communication phase, while integrating early stopping mechanisms further minimizes overall emissions. Conclusion: The proposed GreenDFL provides a comprehensive and practical approach for assessing the sustainability of DFL systems. Furthermore, it offers best practices for improving environmental efficiency in DFL, making sustainability considerations more actionable in real-world deployments.

Artificial intelligence (AI) increasingly influences critical decision-making across sectors. Federated Learning (FL), as a privacy-preserving collaborative AI paradigm, not only enhances data protection but also holds significant promise for intelligent network management, including distributed monitoring, adaptive control, and edge intelligence. Although the trustworthiness of FL systems has received growing attention, the sustainability dimension remains insufficiently explored, despite its importance for scalable real-world deployment. To address this gap, this work introduces sustainability as a distinct pillar within a comprehensive trustworthy FL taxonomy, consistent with AIHLEG guidelines. This pillar includes three key aspects: hardware efficiency, federation complexity, and the carbon intensity of energy sources. Experiments using the FederatedScope framework under diverse scenarios, including varying participants, system complexity, hardware, and energy configurations, validate the practicality of the approach. Results show that incorporating sustainability into FL evaluation supports environmentally responsible deployment, enabling more efficient, adaptive, and trustworthy network services and management AI models.

Federated Learning (FL) is widely recognized as a privacy-preserving Machine Learning paradigm due to its model-sharing mechanism that avoids direct data exchange. Nevertheless, model training leaves exploitable traces that can be used to infer sensitive information. In Decentralized FL (DFL), the topology, defining how participants are connected, plays a crucial role in shaping the model’s privacy, robustness, and convergence. However, the topology introduces an unexplored vulnerability: attackers can exploit it to infer participant relationships and launch targeted attacks. This work uncovers the hidden risks of DFL topologies by proposing a novel Topology Inference Attack that infers the topology solely from model behavior. A taxonomy of topology inference attacks is introduced, categorizing them by the attacker’s capabilities and knowledge. Practical attack strategies are designed for various scenarios, and experiments are conducted to identify key factors influencing attack success. The results demonstrate that analyzing only the model of each node can accurately infer the DFL topology, highlighting a critical privacy risk in DFL systems. These findings offer insights for improving privacy preservation in DFL environments.

Decentralized Federated Learning (DFL) trains models in a collaborative and privacy-preserving manner while removing model centralization risks and improving communication bottlenecks. However, DFL faces challenges in efficient communication management and model aggregation within decentralized environments, especially with heterogeneous data distributions. Thus, this paper introduces ProFe, a novel communication optimization algorithm for DFL that combines knowledge distillation, prototype learning, and quantization techniques. ProFe utilizes knowledge from large local models to train smaller ones for aggregation, incorporates prototypes to better learn unseen classes, and applies quantization to reduce data transmitted during communication rounds. The performance of ProFe has been validated and compared to the literature by using benchmark datasets like MNIST, CIFAR10, and CIFAR100. Results showed that the proposed algorithm reduces communication costs by up to$\approx 40-50 \%$while maintaining or improving model performance. In addition, it adds$\approx 20 \%$training time due to increased complexity, generating a trade-off.

Synthetic data generation, leveraging generative machine learning techniques, offers a promising approach to mitigating privacy concerns associated with real-world data usage. Synthetic data closely resemble real-world data while maintaining strong privacy guarantees. However, a comprehensive assessment framework is still missing in the evaluation of synthetic data generation, especially when considering the balance between privacy preservation and data utility in synthetic data. This research bridges this gap by proposing FEST, a systematic framework for evaluating synthetic tabular data. FEST integrates diverse privacy metrics (attack-based and distance-based), along with similarity and machine learning utility metrics, to provide a holistic assessment. We develop FEST as an open-source Python-based library and validate it on multiple datasets, demonstrating its effectiveness in analyzing the privacy-utility trade-off of different synthetic data generation models. The source code o f FEST is available on Github.

Decentralized Federated Learning (DFL) enables collaborative, privacy-preserving model training without relying on a central server. This decentralized approach reduces bottlenecks and eliminates single points of failure, enhancing scalability and resilience. However, DFL also introduces challenges, such as suboptimal models with non-IID data distributions, increased communication overhead, and resource usage. Thus, this work proposes S-VOTE, a voting-based client selection mechanism that optimizes resource usage and enhances model performance in federations with non-IID data conditions. S-VOTE considers an adaptive strategy for spontaneous local training that addresses participation imbalance, allowing underutilized clients to contribute without significantly increasing resource costs. Extensive experiments on benchmark datasets demonstrate the S-VOTE effectiveness. More in detail, it achieves lower communication costs by up to 21%, 4-6% faster convergence, and improves local performance by 9-17% compared to baseline methods in some configurations, all while achieving a 14-24% energy consumption reduction. These results highlight the potential of S-VOTE to address DFL challenges in heterogeneous environments.

Federated Learning (FL) enables collaborative model training without sharing raw data, preserving participant privacy. Decentralized FL (DFL) eliminates reliance on a central server, mitigating the single point of failure inherent in the traditional FL paradigm, while introducing deployment challenges on resource-constrained devices. To evaluate real-world applicability, this work designs and deploys a physical testbed using edge devices such as Raspberry Pi and Jetson Nano. The testbed is built upon a DFL training platform, NEBULA, and extends it with a power monitoring module to measure energy consumption during training. Experiments across multiple datasets show that model performance is influenced by the communication topology, with denser topologies leading to better outcomes in DFL settings.

Federated Learning (FL), introduced in 2016, was designed to enhance data privacy in collaborative model training environments. Among the FL paradigm, horizontal FL, where clients share the same set of features but different data samples, has been extensively studied in both centralized and decentralized settings. In contrast, Vertical Federated Learning (VFL), which is crucial in real-world decentralized scenarios where clients possess different, yet sensitive, data about the same entity, remains underexplored. Thus, this work introduces De-VertiFL, a novel solution for training models in a decentralized VFL setting. De-VertiFL contributes by introducing a new network architecture distribution, an innovative knowledge exchange scheme, and a distributed federated training process. Specifically, De-VertiFL enables the sharing of hidden layer outputs among federation clients, allowing participants to benefit from intermediate computations, thereby improving learning efficiency. De-VertiFL has been evaluated using a variety of well-known datasets, including both image and tabular data, across binary and multiclass classification tasks. The results demonstrate that De-VertiFL generally surpasses state-of-the-art methods in F1-score performance, while maintaining a decentralized and privacy-preserving framework.

The metaverse has gained tremendous popularity in recent years, allowing the interconnection of users worldwide. However, current systems in metaverse scenarios, such as virtual reality glasses, offer a partial immersive experience. In this context, Brain–Computer Interfaces (BCIs) can introduce a revolution in the metaverse, although a study of the applicability and implications of BCIs in these virtual scenarios is required. Based on a limited number of publications, this work reviews the applicability of BCIs in the metaverse, analyzing the current status of this integration based on different categories related to virtual worlds and the evolution of BCIs in these scenarios in the medium and long term. This work also proposes the design and implementation of a general framework that integrates BCIs with different data sources from sensors and actuators (e.g., VR glasses) based on a modular design to be easily extended. This manuscript also validates the framework in a demonstrator consisting of driving a car within a metaverse, using a BCI for neural data acquisition, a VR headset to provide realism, and a steering wheel and pedals. Four use cases (UCs) are selected, focusing on cognitive and emotional assessment of the driver, detection of drowsiness, and driver authentication while using the vehicle. The results demonstrate the applicability of BCIs to metaverse scenarios using the proposed framework, achieving over 80% F1-score for all UCs, with performance close to 100% for detecting emotions and authenticating users. Moreover, this manuscript offers an analysis of BCI trends in the metaverse, also identifying future challenges that the intersection of these technologies will face. Finally, it reviews the concerns that using BCIs in virtual world applications could generate according to different categories: accessibility, user inclusion, privacy, cybersecurity, physical safety, and ethics.

Brain-Computer Interfaces (BCIs) are systems traditionally used in medicine and designed to interact with the brain to record or stimulate neurons. Despite their benefits, the literature has demonstrated that invasive BCIs focused on neurostimulation present vulnerabilities allowing attackers to gain control. In this context, neural cyberattacks emerged as threats able to disrupt spontaneous neural activity by performing neural overstimulation or inhibition. Previous work validated these attacks in small-scale simulations with a reduced number of neurons, lacking real-world complexity. Thus, this work tackles this limitation by analyzing the impact of two existing neural attacks, Neuronal Flooding (FLO) and Neuronal Jamming (JAM), on a complex neuronal topology of the primary visual cortex of mice consisting of approximately 230,000 neurons, tested on three realistic visual stimuli: flash effect, movie, and drifting gratings. Each attack was evaluated over three relevant events per stimulus, also testing the impact of attacking 25 % and 50 % of the neurons. The results, based on the number of spikes and shift percentage metrics, showed that the attacks caused the greatest impact on the movie, while dark and static events exhibited highest resilience. Although both attacks can significantly affect neural activity, JAM was generally more damaging, producing longer temporal delays, and had a larger prevalence. Finally, JAM did not require altering many neurons to significantly affect neural activity, while the impact of FLO increased with the number of neurons attacked. • Evaluation of two neural cyberattacks on a realistic simulation of the visual cortex. • Use of a rich topology with 230,000 neurons from six cortical layers. • Attacks tested on different events within three visual stimuli used as visual input. • Evaluation of both threats using two metrics: number of spikes and shift percentage. • Both attacks can disrupt neural activity, having effects lasting for hundreds of milliseconds.

Although ransomware has received broad attention in media and research, this evolving threat vector still poses a systematic threat. Related literature has explored their detection using various approaches leveraging Machine and Deep Learning. While these approaches are effective in detecting malware, they do not answer how to use this intelligence to protect against threats, raising concerns about their applicability in a hostile environment. Solutions that focus on mitigation rarely explore how to prevent and not just alert or halt its execution, especially when considering Linux-based samples. This paper presents GuardFS , a file system-based approach to investigate the integration of detection and mitigation of ransomware. Using a bespoke overlay file system, data is extracted before files are accessed. Models trained on this data are used by three novel defense configurations that obfuscate, delay, or track access to the file system. The experiments on GuardFS test the configurations in a reactive setting. The results demonstrate that although data loss cannot be completely prevented, it can be significantly reduced. Usability and performance analysis demonstrate that the defense effectiveness of the configurations relates to their impact on resource consumption and usability.

Recent research has shown that the integration of Reinforcement Learning (RL) with Moving Target Defense (MTD) can enhance cybersecurity in Internet-of-Things (IoT) devices. Nevertheless, the practicality of existing work is hindered by data privacy concerns associated with centralized data processing in RL, and the unsatisfactory time needed to learn right MTD techniques that are effective against a rising number of heterogeneous zero-day attacks. Thus, this work presents CyberForce, a framework that combines Federated and Reinforcement Learning (FRL) to collaboratively and privately learn suitable MTD techniques for mitigating zero-day attacks. CyberForce integrates device fingerprinting and anomaly detection to reward or penalize MTD mechanisms chosen by an FRL-based agent. The framework has been deployed and evaluated in a scenario consisting of ten physical devices of a real IoT platform affected by heterogeneous malware samples. A pool of experiments has demonstrated that CyberForce learns the MTD technique mitigating each attack faster than existing RL-based centralized approaches. In addition, when various devices are exposed to different attacks, CyberForce benefits from knowledge transfer, leading to enhanced performance and reduced learning time in comparison to recent works. Finally, different aggregation algorithms used during the agent learning process provide CyberForce with notable robustness to malicious attacks.

Nowadays, the usage of computers ranges from activities that do not consider sensitive data, such as playing video games, to others managing confidential information, like military operations. Additionally, regardless of the actions performed by subjects, most computers store different pieces of sensitive data, making the implementation of robust security mechanisms a critical and mandatory task. In this context, continuous authentication has been proposed as a complementary mechanism to improve the limitations of conventional authentication methods. However, mainly driven by the evolution of Machine Learning (ML), a series of challenges related to authentication performance and, therefore, the feasibility of existing systems are still open. This work proposes the usage of contextual information related to the applications executed in the computers to create ML models able to authenticate subjects continuously. To evaluate the suitability of the proposed context-aware ML models, a continuous authentication framework for computers has been designed and implemented. Then, a set of experiments with a public dataset with 12 subjects demonstrated the improvement of the proposed approach compared to the existing ones. Precision, recall, and F1-Score metrics are raised from an average of 0.96 (provided by general ML models proposed in the literature) to 0.99-1.

Decentralized Federated Learning (DFL), a paradigm for managing big data in a privacy-preserving and distributed manner, is vulnerable to poisoning attacks where malicious clients tamper with data or models. Current defense methods often assume Independently and Identically Distributed (IID) data across participants, which is unrealistic in real-world applications. In more realistic non-IID contexts, existing defensive strategies face challenges when distinguishing between models that have been compromised and those that have been trained on heterogeneous data distributions (non-IID), leading to diminished efficacy. In response, this paper proposes a framework that employs the Moving Target Defense (MTD) approach to bolster the robustness of DFL models. By continuously modifying the attack surface of the DFL system, the framework aims to mitigate poisoning attacks effectively. The proposed solution includes both proactive and reactive modes, utilizing a reputation system that combines metrics of model similarity and loss, alongside various defensive techniques. Comprehensive experimental evaluations indicate that the MTD-based mechanism significantly mitigates a range of poisoning attack types across multiple datasets with different federation topologies.

Synthetic data generation has emerged as a powerful solution to meet the demand for high-quality, diverse, and privacy-preserving data in many domains. Still, there is an open challenge when dealing with class imbalance and privacy preservation in synthetic tabular data generation. Thus, this study introduces two algorithms: balanced Tabular Generative Adversarial Network (b-TGAN) and balanced Tabular Principle Component Analysis (b-TPCA). While b-TGAN proactively tackles class imbalance by incorporating a re-balancing mechanism and leveraging an Autoencoder, b-TPCA offers a privacy-preserving solution by generating synthetic data using statistical properties. Through experiments on five datasets, this study demonstrates the effectiveness of b-TGAN in generating balanced data, particularly in improving the performance on minority classes. b-TPCA also shows promising results, achieving comparable ML utility to the baseline method while enhancing privacy preservation.

Artificial Intelligence (AI) is a rapidly integrated technology, significantly contributing to advancements like 6G. However, its swift adoption raises considerable security concerns. Large Language Models (LLMs) pose risks such as spear phishing, code injections, and remote code execution. Conventional threat modeling, used in secure software development, faces challenges when applied to AI systems, as existing methodologies are designed for traditional software. Furthermore, AI-specific threat modeling research is sparse and lacks approaches providing practical support or automation. Thus, this demo paper presents ThreatFinderAI, an asset-centric threat modeling and risk assessment framework. ThreatFinderAI fulfills seven steps aligned with AI system design and transforms AI threat and control knowledge bases into a queryable knowledge graph for automated asset identification and threat elicitation. It also proposes business impact analysis and expert estimates for AI threat impact quantification. In the demonstration, ThreatFinderAI is illustrated by securing a customer care application relying on LLMs. Through this, it is demonstrated how the proposed framework can be used to identify relevant threats and practical countermeasures and communicate strategic risk.

Decentralized Federated Learning (DFL) emerges as an innovative paradigm to train collaborative models, addressing the single point of failure limitation. However, the security and trustworthiness of FL and DFL are compromised by poisoning attacks, negatively impacting its performance. Existing defense mechanisms have been designed for centralized FL and they do not adequately exploit the particularities of DFL. Thus, this work introduces Sentinel, a defense strategy to counteract poisoning attacks in DFL. Sentinel leverages the accessibility of local data and defines a three-step aggregation protocol consisting of similarity filtering, bootstrap validation, and normalization to safeguard against malicious model updates. Sentinel has been evaluated with diverse datasets and data distributions. Besides, various poisoning attack types and threat levels have been verified. The results improve the state-of-the-art performance against both untargeted and targeted poisoning attacks when data follows an IID (Independent and Identically Distributed) configuration. Besides, under non-IID configuration, it is analyzed how performance degrades both for Sentinel and other state-of-the-art robust aggregation methods.

In the era of the Internet of Things and Artificial Intelligence (AI), cybersecurity has become a critical challenge. Existing AI-based detection systems have shown merit in detecting heterogeneous malware, but their effectiveness against the next generation of AI-powered ransomware encrypting data intelligently is under discussion. In this context, this work introduces a novel AI-powered ransomware that combines Reinforcement Learning and fingerprints based on system calls to evade AI-based cybersecurity detection systems. More in detail, an RL-based agent utilizes Deep Q-Learning as a learning algorithm, system calls to model device states, and the result of unsupervised learning as input for the reward function. The efficacy of this approach has been evaluated on a Raspberry Pi acting as a real spectrum sensor that hosts a behavioral AI-based anomaly detector. Experiments have compared the AI-based ransomware evasion performance with the literature and analyzed how various benign and realistic behaviors affect evasion performance. Results showed that precise and adaptive evasion capabilities can be learned in a few minutes, motivating the need for better cybersecurity systems.

Ransomware has remained one of the most notorious threats in the cybersecurity field, for which Moving Target Defense (MTD) has been proposed as a novel defense paradigm. Although various approaches leverage MTD, few of them rely on the operating system and, specifically, the file system, thereby making them dependent on other computing devices, rendering defense against certain threats unrealistic. File-based approaches are less studied here while showing limitations in resource usage and defense effectiveness. Furthermore, existing ransomware defenses merely restore data or detect attacks without preventing them. Thus, this paper introduces the MTFS file system and the design and implementation of three novel MTD techniques – one delaying attackers, one trapping recursive directory traversal, and another one hiding file types. The effectiveness of the techniques is shown in three experiments. First, it is demonstrated that the techniques can delay and mitigate ransomware on real IoT devices. Secondly, in a broader scope, the solution was confronted with 13 ransomware samples, highlighting that it can save 97% of the files. Regarding overhead, the defense system consumes only a small amount of resources, highlighting the feasibility of proactive defense.

The growing concern over malicious attacks targeting the robustness of both Centralized and Decentralized Federated Learning (FL) necessitates novel defensive strategies. In contrast to the centralized approach, Decentralized FL (DFL) has the advantage of utilizing network topology and local dataset information, enabling the exploration of Moving Target Defense (MTD) based approaches.This work presents a theoretical analysis of the influence of network topology on the robustness of DFL models. Drawing inspiration from these findings, a three-stage MTD-based aggregation protocol, called Voyager, is proposed to improve the robustness of DFL models against poisoning attacks by manipulating network topology connectivity. Voyager has three main components: an anomaly detector, a network topology explorer, and a connection deployer. When an abnormal model is detected in the network, the topology explorer responds strategically by forming connections with more trustworthy participants to secure the model. Experimental evaluations show that Voyager effectively mitigates various poisoning attacks without imposing significant resource and computational burdens on participants. These findings highlight the proposed reactive MTD as a potent defense mechanism in the context of DFL.

Abstract Federated learning (FL) enables participants to collaboratively train machine and deep learning models while safeguarding data privacy. However, the FL paradigm still has drawbacks that affect its trustworthiness, as malicious participants could launch adversarial attacks against the training process. Previous research has examined the robustness of horizontal FL scenarios under various attacks. However, there is a lack of research evaluating the robustness of decentralized vertical FL and comparing it with horizontal FL architectures affected by adversarial attacks. Therefore, this study proposes three decentralized FL architectures: HoriChain, VertiChain, and VertiComb. These architectures feature different neural networks and training protocols suitable for horizontal and vertical scenarios. Subsequently, a decentralized, privacy-preserving, and federated use case with non-IID data to classify handwritten digits is deployed to assess the performance of the three architectures. Finally, a series of experiments computes and compares the robustness of the proposed architectures when they are affected by different data poisoning methods, including image watermarks and gradient poisoning adversarial attacks. The experiments demonstrate that while specific configurations of both attacks can undermine the classification performance of the architectures, HoriChain is the most robust one.

The proliferation of the Internet of Things (IoT) has led to the emergence of crowdsensing applications, where a multitude of interconnected devices collaboratively collect and analyze data. Ensuring the authenticity and integrity of the data collected by these devices is crucial for reliable decision-making and maintaining trust in the system. Traditional authentication methods are often vulnerable to attacks or can be easily duplicated, posing challenges to securing crowdsensing applications. Besides, current solutions leveraging device behavior are mostly focused on device identification, which is a simpler task than authentication. To address these issues, an individual IoT device authentication framework based on hardware behavior fingerprinting and Transformer autoencoders is proposed in this work. To support the design, a threat model details the security problems faced when performing hardware-based authentication in IoT. This solution leverages the inherent imperfections and variations in IoT device hardware to differentiate between devices with identical specifications. By monitoring and analyzing the behavior of key hardware components, such as the CPU, GPU, RAM, and Storage on devices, unique fingerprints for each device are created. The performance samples are considered as time series data and used to train outlier detection transformer models, one per device and aiming to model its normal data distribution. Then, the framework is validated within a spectrum crowdsensing system leveraging Raspberry Pi devices. After a pool of experiments, the model from each device is able to individually authenticate it between the 45 devices employed for validation. An average True Positive Rate (TPR) of 0.74±0.13 and an average maximum False Positive Rate (FPR) of 0.06±0.09 demonstrate the effectiveness of this approach in enhancing authentication, security, and trust in crowdsensing applications.

In 2016, Google proposed Federated Learning (FL) as a novel paradigm to train Machine Learning (ML) models across the participants of a federation while preserving data privacy. Since its birth, Centralized FL (CFL) has been the most used approach, where a central entity aggregates participants’ models to create a global one. However, CFL presents limitations such as communication bottlenecks, single point of failure, and reliance on a central server. Decentralized Federated Learning (DFL) addresses these issues by enabling decentralized model aggregation and minimizing dependency on a central entity. Despite these advances, current platforms training DFL models struggle with key issues such as managing heterogeneous federation network topologies, adapting the FL process to virtualized or physical deployments, and using a limited number of metrics to evaluate different federation scenarios for efficient implementation. To overcome these challenges, this paper presents Fedstellar, a novel platform designed to train FL models in a decentralized, semi-decentralized, and centralized fashion across diverse federations of physical or virtualized devices. Fedstellar allows users to create federations by customizing parameters like the number and type of devices training FL models, the network topology connecting them, the machine and deep learning algorithms, or the datasets of each participant, among others. Additionally, it offers real-time monitoring of model and network performance. The Fedstellar implementation encompasses a web application with an interactive graphical interface, a controller for deploying federations of nodes using physical or virtual devices, and a core deployed on each device, which provides the logic needed to train, aggregate, and communicate in the network. The effectiveness of the platform has been demonstrated in two scenarios: a physical deployment involving single-board devices such as Raspberry Pis for detecting cyberattacks and a virtualized deployment comparing various FL approaches in a controlled environment using MNIST and CIFAR-10 datasets. In both scenarios, Fedstellar demonstrated consistent performance and adaptability, achieving F1scores of 91%, 98%, and 91.2% using DFL for detecting cyberattacks and classifying MNIST and CIFAR-10, respectively, reducing training time by 32% compared to centralized approaches.

In the last years, the number of IoT devices deployed has suffered an undoubted explosion, reaching the scale of billions. However, some new cybersecurity issues have appeared together with this development. Some of these issues are the deployment of unauthorized devices, malicious code modification, malware deployment, or vulnerability exploitation. This fact has motivated the requirement for new device identification mechanisms based on behavior monitoring. Besides, these solutions have recently leveraged Machine and Deep Learning (ML/DL) techniques due to the advances in this field and the increase in processing capabilities. In contrast, attackers do not stay stalled and have developed adversarial attacks focused on context modification and ML/DL evaluation evasion applied to IoT device identification solutions. However, literature has not yet analyzed in detail the impact of these attacks on individual identification solutions and their countermeasures. This work explores the performance of hardware behavior-based individual device identification, how it is affected by possible context- and ML/DL-focused attacks, and how its resilience can be improved using defense techniques. In this sense, it proposes an LSTM-CNN architecture based on hardware performance behavior for individual device identification. Then, the most usual ML/DL classification techniques have been compared with the proposed architecture using a hardware performance dataset collected from 45 Raspberry Pi devices running identical software. The LSTM-CNN improves previous solutions achieving a ＋0.96 average F1-Score and 0.8 minimum TPR for all devices. Afterward, context- and ML/DL-focused adversarial attacks were applied against the previous model to test its robustness. A temperature-based context attack was not able to disrupt the identification, but some ML/DL state-of-the-art evasion attacks were successful. Finally, adversarial training and model distillation defense techniques are selected to improve the model resilience to evasion attacks, improving its robustness from up to 0.88 attack success ratio to 0.17 in the worst attack case, without degrading its performance in an impactful manner.

The rapid expansion of the Internet of Things (IoT) and Edge Computing has presented challenges for centralized Machine and Deep Learning (ML/DL) methods due to the presence of distributed data silos that hold sensitive information. To address concerns regarding data privacy, collaborative and privacy-preserving ML/DL techniques like Federated Learning (FL) have emerged. FL ensures data privacy by design, as the local data of participants remains undisclosed during the creation of a global and collaborative model. However, data privacy and performance are insufficient since a growing need demands trust in model predictions. Existing literature has proposed various approaches dealing with trustworthy ML/DL (excluding data privacy), identifying robustness, fairness, explainability, and accountability as important pillars. Nevertheless, further research is required to identify trustworthiness pillars and evaluation metrics specifically relevant to FL models, as well as to develop solutions that can compute the trustworthiness level of FL models. This work examines the existing requirements for evaluating trustworthiness in FL and introduces a comprehensive taxonomy consisting of six pillars (privacy, robustness, fairness, explainability, accountability, and federation), along with over 30 metrics for computing the trustworthiness of FL models. Subsequently, an algorithm named FederatedTrust is designed based on the pillars and metrics identified in the taxonomy to compute the trustworthiness score of FL models. A prototype of FederatedTrust is implemented and integrated into the learning process of FederatedScope, a well-established FL framework. Finally, five experiments are conducted using different configurations of FederatedScope (with different participants, selection rates, training rounds, and differential privacy) to demonstrate the utility of FederatedTrust in computing the trustworthiness of FL models. Three experiments employ the FEMNIST dataset, and two utilize the N-BaIoT dataset, considering a real-world IoT security use case.

Brain–Computer Interfaces (BCIs) allow the acquisition of brain activity using non-invasive techniques such as Electroencephalography (EEG). Since BCI devices do not commonly interpret the acquired EEG signals, external software applications play a critical role in the BCI lifecycle. Despite the literature offering a great variety of platforms and frameworks, they present several limitations, such as implementing old software architectures or needing more functionality to cover all phases of the BCI lifecycle. Based on these limitations, this work proposes the design and implementation of NeuronLab, a secure, multi-platform, standalone, multi-paradigm, and web-based framework that defines all BCI lifecycle phases and provides novel functionality compared to current open-source BCI software, such as sharing experiments between researchers and storing data on the cloud. This framework has been validated in two experiments common in BCI literature: P300 identification and limb movements detection. This verification has been performed based on performance metrics, such as CPU and RAM consumption, highlighting that NeuronLab is a promising solution for BCI scenarios requiring a distributed and collaborative platform for researchers and practitioners.

Abstract In response to the global safety concern of drowsiness during driving, the European Union enforces that new vehicles must integrate detection systems compliant with the general data protection regulation. To identify drowsiness patterns while preserving drivers’ data privacy, recent literature has combined Federated Learning (FL) with different biosignals, such as facial expressions, heart rate, electroencephalography (EEG), or electrooculography (EOG). However, existing solutions are unsuitable for drowsiness detection where heterogeneous stakeholders want to collaborate at different levels while guaranteeing data privacy. There is a lack of works evaluating the benefits of using Hierarchical FL (HFL) with EEG and EOG biosignals, and comparing HFL over traditional FL and Machine Learning (ML) approaches to detect drowsiness at the wheel while ensuring data confidentiality. Thus, this work proposes a flexible framework for drowsiness identification by using HFL, FL, and ML over EEG and EOG data. To validate the framework, this work defines a scenario of three transportation companies aiming to share data from their drivers without compromising their confidentiality, defining a two-level hierarchical structure. This study presents three incremental Use Cases (UCs) to assess detection performance: UC1) intra-company FL, yielding a 77.3% accuracy while ensuring the privacy of individual drivers’ data; UC2) inter-company FL, achieving 71.7% accuracy for known drivers and 67.1% for new subjects, ensuring data confidentiality between companies but not intra-organization; and UC3) HFL inter-company, which ensured comprehensive data privacy both within and between companies, with an accuracy of 71.9% for training subjects and 65.5% for new subjects.

Integrated sensing and communication is a novel paradigm using crowdsensing spectrum sensors to help with the management of spectrum scarcity. However, well-known vulnerabilities of resource-constrained spectrum sensors and the possibility of being manipulated by users with physical access complicate their protection against spectrum sensing data falsification (SSDF) attacks. Most recent literature suggests using behavioral fingerprinting and Machine/Deep Learning (ML/DL) for improving similar cybersecurity issues. Nevertheless, the applicability of these techniques in resource-constrained devices, the impact of attacks affecting spectrum data integrity, and the performance and scalability of models suitable for heterogeneous sensors types are still open challenges. To improve limitations, this work presents seven SSDF attacks affecting spectrum sensors and introduces CyberSpec, an ML/DL-oriented framework using device behavioral fingerprinting to detect anomalies produced by SSDF attacks. CyberSpec has been implemented and validated in ElectroSense, a real crowdsensing RF monitoring platform where several configurations of the proposed SSDF attacks have been executed in different sensors. A pool of experiments with different unsupervised ML/DL-based models has demonstrated the suitability of CyberSpec detecting the previous attacks within an acceptable timeframe.

Device fingerprinting combined with Machine and Deep Learning (ML/DL) report promising performance when detecting spectrum sensing data falsification (SSDF) attacks. However, the amount of data needed to train models and the scenario privacy concerns limit the applicability of centralized ML/DL. Federated learning (FL) addresses these drawbacks but is vulnerable to adversarial participants and attacks. The literature has proposed countermeasures, but more effort is required to evaluate the performance of FL detecting SSDF attacks and their robustness against adversaries. Thus, the first contribution of this work is to create an FL-oriented dataset modeling the behavior of resource-constrained spectrum sensors affected by SSDF attacks. The second contribution is a pool of experiments analyzing the robustness of FL models according to i) three families of sensors, ii) eight SSDF attacks, iii) four FL scenarios dealing with anomaly detection and binary classification, iv) up to 33% of participants implementing data and model poisoning attacks, and v) four aggregation functions acting as anti-adversarial mechanisms. In conclusion, FL achieves promising performance when detecting SSDF attacks. Without anti-adversarial mechanisms, FL models are particularly vulnerable with$>$16% of adversaries. Coordinate-wise-median is the best mitigation for anomaly detection, but binary classifiers are still affected with$>$33% of adversaries.

Moving Target Defense (MTD) is a promising approach to mitigate attacks by dynamically altering target attack surfaces. Still, selecting suitable MTD techniques for zero-day attacks is an open challenge. Reinforcement Learning (RL) could be an effective approach to optimize the MTD selection through trial and error, but the literature fails when i) evaluating the performance of RL and MTD solutions in real-world scenarios, ii) studying whether behavioral fingerprinting is suitable for RL, and iii) calculating the consumption of resources in single-board computers (SBC). Thus, the work at hand proposes an online RL-based framework that learns correct MTD mechanisms mitigating heterogeneous zero-day attacks in SBC. The framework considers behavioral fingerprinting to represent SBCs’ states and RL to learn MTD techniques that mitigate each malicious state. It has been deployed on a real IoT crowdsensing scenario with a Raspberry Pi acting as a spectrum sensor. The Raspberry Pi has been infected with different samples of command and control malware, rootkits, and ransomware to later select between four existing MTD techniques. A set of experiments demonstrated the suitability of the framework to learn proper MTD techniques mitigating all attacks (except a harmfulness rootkit) while consuming$\approx 10$% of RAM, and negligible CPU.

Abstract The rise of Decentralized Federated Learning (DFL) has enabled the training of machine learning models across federated participants, fostering decentralized model aggregation and reducing dependence on a server. However, this approach introduces unique communication security challenges that have yet to be thoroughly addressed in the literature. These challenges primarily originate from the decentralized nature of the aggregation process, the varied roles and responsibilities of the participants, and the absence of a central authority to oversee and mitigate threats. Addressing these challenges, this paper first delineates a comprehensive threat model focused on DFL communications. In response to these identified risks, this work introduces a security module to counter communication-based attacks for DFL platforms. The module combines security techniques such as symmetric and asymmetric encryption with Moving Target Defense (MTD) techniques, including random neighbor selection and IP/port switching. The security module is implemented in a DFL platform, Fedstellar, allowing the deployment and monitoring of the federation. A DFL scenario with physical and virtual deployments have been executed, encompassing three security configurations: (i) a baseline without security, (ii) an encrypted configuration, and (iii) a configuration integrating both encryption and MTD techniques. The effectiveness of the security module is validated through experiments with the MNIST dataset and eclipse attacks.The results showed an average F1 score of 95%, with the most secure configuration resulting in CPU usage peaking at 68% (± 9%) in virtual deployments and network traffic reaching 480.8 MB (± 18 MB), effectively mitigating risks associated with eavesdropping or eclipse attacks.

Abstract Brain-computer interfaces (BCIs) are widely used in medical scenarios to treat neurological conditions, such as Parkinson’s disease or epilepsy, when a pharmacological approach is ineffective. Despite their advantages, these BCIs target relatively large areas of the brain, causing side effects. In this context, projects such as Neuralink aim to stimulate and inhibit neural activity with single-neuron resolution, expand their usage to other sectors, and thus democratize access to neurotechnology. However, these initiatives present vulnerabilities in their designs that cyberattackers can exploit to cause brain damage. Specifically, the literature has documented the applicability of neural cyberattacks, threats capable of stimulating or inhibiting individual neurons to alter spontaneous neural activity. However, these works were limited by a lack of realistic neuronal topologies to test the cyberattacks. Surpassed this limitation, this work considers a realistic neuronal representation of the primary visual cortex of mice to evaluate the impact of neural cyberattacks more realistically. For that, this publication evaluates two existing cyberattacks, Neuronal Flooding and Neuronal Jamming, assessing the impact that different voltages on a particular set of neurons and the number of neurons simultaneously under attack have on the amount of neural activity produced. As a result, both cyberattacks increased the number of neural activations, propagating their impact for approximately 600 ms, where the activity converged into spontaneous behavior. These results align with current evidence about the brain, highlighting that neurons will tend to their baseline behavior after the attack.

Cybersecurity solutions have shown promising performance when detecting ransomware samples that use fixed algorithms and encryption rates. However, due to the current explosion of Artificial Intelligence (AI), sooner than later, ransomware, and malware in general, will incorporate AI techniques to intelligently and dynamically adapt its behavior to be undetected. It might result in ineffective and obsolete cybersecurity solutions, but the literature lacks AI-powered ransomware samples to verify it. Thus, this work proposes RansomAI, a Reinforcement Learning-based framework that can be integrated into existing ransomware samples to adapt their encryption behavior and stay stealthy while encrypting files. RansomAI presents an agent that learns the best encryption algorithm, rate, and duration that minimizes its detection (using a reward mechanism and a fingerprinting intelligent detection system) while maximizing its damage. The proposed framework was validated with Ransomware-PoC, a ransomware that infected a Raspberry Pi 4 acting as a crowdsensor. A pool of experiments with Deep Q-Learning and Isolation Forest (deployed on the agent and detection system, respectively) has demonstrated that RansomAI evades the detection of Ransomware-PoC affecting the Raspberry Pi 4 in a few minutes with >90% accuracy.

Malware affecting Internet of Things (IoT) devices is rapidly growing due to the relevance of this paradigm in real-world scenarios. Specialized literature has also detected a trend towards multi-purpose malware able to execute different malicious actions such as remote control, data leakage, encryption, or code hiding, among others. Protecting IoT devices against this kind of malware is challenging due to their well-known vulnerabilities and limitation in terms of CPU, memory, and storage. To improve it, the moving target defense (MTD) paradigm was proposed a decade ago and has shown promising results, but there is a lack of IoT MTD solutions dealing with multi-purpose malware. Thus, this work proposes four MTD mechanisms changing IoT devices' network, data, and runtime environment to mitigate multi-purpose malware. Furthermore, it presents a lightweight and IoT-oriented MTD framework to decide what, when, and how the MTD mechanisms are deployed. Finally, the efficiency and effectiveness of the framework and MTD mechanisms are evaluated in a real-world scenario with one IoT spectrum sensor affected by multi-purpose malware.

This paper presents Fedstellar, a platform for training decentralized Federated Learning (FL) models in heterogeneous topologies in terms of the number of federation participants and their connections. Fedstellar allows users to build custom topologies, enabling them to control the aggregation of model parameters in a decentralized manner. The platform offers a Web application for creating, managing, and connecting nodes to ensure data privacy and provides tools to measure, monitor, and analyze the performance of the nodes. The paper describes the functionalities of Fedstellar and its potential applications. To demonstrate the applicability of the platform, different use cases are presented in which decentralized, semi-decentralized, and centralized architectures are compared in terms of model performance, convergence time, and network overhead when collaboratively classifying hand-written digits using the MNIST dataset.

Cybersecurity solutions based on machine learning (ML) and behavioral fingerprinting have demonstrated their suitability when detecting heterogeneous malware. However, most solutions are black boxes missing explainable and visual capabilities needed to analyze relevant metrics and malicious behaviors to be collected. In this demonstration, SecBox, a dynamic malware analysis platform with integrated data collection and visualization for malware execution, is presented. To provide a lightweight sandboxing approach, the architecture relies on Linux containers for isolation. The sandboxing and data analysis components of the SecBox architecture are deployed in a test bed to show the analysis of two malware families. In the presented scenario, the Monti ransomware and CoinMiner, a Monero-based cryptojacker are analyzed after obtaining them from a public database.

Traditionally, IoT crowdsensing devices have been outside the cryptomining domain due to their limitations in terms of computational power. In 2014, Monero (XNR) changed this situation forever. Monero is an open-source digital payment token that can be mined in resource-constrained devices like IoT and single-board computers. Despite the Monero advantages, it opened the door for cryptojackers illicitly mining cryptocurrencies by exploiting well-known vulnerabilities of IoT devices. Existing detection solutions provide good performance while detecting the mining phase of cryptojackers, but early detection is desired to avoid malware spreading and resource misuse. Thus, this work proposes a framework that combines device behavioral fingerprinting and machine learning to detect and classify preparatory phases of cryptojackers. The framework has been deployed in a crowdsensing IoT spectrum sensor, Raspberry Pi, infected by a recent cryptojacker called Linux.MulDrop.14. Promising detection results demonstrate the framework’s suitability while detecting early phases of cryptojackers.

Trust, security, and privacy are three of the major pillars to assemble the fifth-generation network and beyond. Despite such pillars are principally interconnected, a multitude of challenges arise that need to be addressed separately. 5G networks ought to offer flexible and pervasive computing capabilities across multiple domains according to user demands and assure trustworthy network providers. To this end, distributed marketplaces expect to boost the trading of heterogeneous resources so as to enable the establishment of pervasive service chains between cross-domains. Yet, the need for selecting reliable parties as “marketplace operators” plays a pivotal role in achieving a trustworthy ecosystem. Two of the principal blockages in managing foreseeable networks are the need to consider trust as a property in the resource provisioning process and adapt previous trust models to accomplish the new network and business requirements. In this regard, this article is centered on the trust management of 5G multi-party network resource provisioning. As a result, a reputation-based trust framework is proposed as a Trust-as-a-Service (TaaS) solution for a distributed multi-stakeholder environment where requirements such as zero trust and zero-touch principles should be met. Besides, a literature review is also conducted to recognize the network and business requirements currently envisaged. Finally, the validation of the proposed trust framework was performed in a real research environment, the 5GBarcelona testbed, leveraging 12% of a 2.1 GHz CPU with 20 cores and 2% of the 30 GiB memory. These outcomes reveal the TaaS solution’s feasibility and conservative approach in the context of determining reliable network operators.

The Internet of Things (IoT), a network of interconnected devices, has grown and gained traction over the last few years. This paradigm can impact our lives while also providing significant economic benefits. However, although resource-constrained IoT devices offer numerous advantages, they are also vulnerable to cyberattacks. As a result, ransomware severely threatens IoT devices managing sensitive and relevant information. Solutions based on Machine and Deep Learning (ML/DL) that consider behavioral data have been identified as promising. However, most detection solutions have been developed for Windows-based systems, which generally have more resources than IoT devices. As a result, these solutions are not suitable for resource-constrained components. In addition, no solution compares the pros and cons of different behavioral dimensions of resource-constrained devices. Thus, this work presents a framework that combines three different behavioral sources with supervised and unsupervised ML/DL algorithms to detect and classify heterogeneous ransomware impacting resource-constrained spectrum sensors. A pool of experiments has demonstrated the suitability of the proposed solution and compared its performance with a rule-based system. In conclusion, the usage of resources combined with local outlier factor and decision tree are the most promising combinations to detect anomalies and classify ransomware while consuming CPU, RAM, and time of devices in a reduced manner.

Abstract When detecting cyberattacks in Industrial settings, it is not sufficient to determine whether the system is suffering a cyberattack. It is also fundamental to explain why the system is under a cyberattack and which are the assets affected. In this context, the Anomaly Detection based on Machine Learning (ML) and Deep Learning (DL) techniques showed great performance when detecting cyberattacks in industrial scenarios. However, two main limitations hinder using them in a real environment. Firstly, most solutions are trained using a supervised approach, which is impractical in the real industrial world. Secondly, the use of black‐box ML and DL techniques makes it impossible to interpret the decision made by the model. This article proposes an interpretable and semi‐supervised system to detect cyberattacks in Industrial settings. Besides, our proposal was validated using data collected from the Tennessee Eastman Process. To the best of our knowledge, this system is the only one that offers interpretability together with a semi‐supervised approach in an industrial setting. Our system discriminates between causes and effects of anomalies and also achieved the best performance for 11 types of anomalies out of 20 with an overall recall of 0.9577, a precision of 0.9977, and a F1‐score of 0.9711.

Crowdsensing platforms collect, process, transmit, and analyze spectrum data worldwide to optimize radio frequency spectrum usage. However, Internet of Things (IoT) spectrum sensors, performing some of the previous tasks, are exposed to software manipulation aiming to execute spectrum sensing data falsification (SSDF) attacks to compromise data integrity and spectrum optimization. Novel intrusion detection systems (IDSs) combining device fingerprinting with machine and deep learning (ML/DL) improve the limitation of traditional solutions and remove the necessity of redundant sensors and reputation mechanisms. However, they fail when detecting SSDF attacks accurately while protecting sensors privacy. This work proposes a novel host-based and federated learning-oriented IDS for IoT spectrum sensors that consider unsupervised ML/DL and fingerprints based on system calls. The framework detection performance and consumption of resources are analyzed in local and federated scenarios with six spectrum sensors deployed on Raspberry Pis. The obtained results significantly improve related work when detecting SSDF attacks while protecting sensors privacy, and consuming CPU, memory, and storage of sensors in a reduced manner.

In the realm of industrial anomaly detection, machine and deep learning models face a critical vulnerability to adversarial attacks. In this context, existing attack methodologies primarily target continuous features, often in the context of images, making them unsuitable for the categorical or discrete features prevalent in industrial systems. To fortify the cybersecurity of industrial environments, this paper introduces a groundbreaking adversarial attack approach tailored to the unique demands of these settings. Our novel technique enables the creation of targeted adversarial samples that are valid within the framework of supervised cyberattack detection models in industrial scenarios, preserving the consistency of discrete values and correcting cases where an adversarial sample transitions into a normal one. Our approach leverages the SHAP interpretability method to identify the most salient features for each sample. Subsequently, the Projected Gradient Descent technique is employed to perturb continuous features, ensuring adversarial sample generation. To handle categorical features for a specific adversarial sample, our method scrutinizes the closest sample within the normal training dataset and replicates its categorical feature values. Additionally, Decision Trees trained within a Random Forest are utilized to ensure that the resulting adversarial samples maintain the essential abnormal behavior required for detection. The validation of our proposal was conducted using the WADI dataset obtained from a water distribution plant, providing a realistic industrial context. During validation, we assessed the mean error and the total number of adversarial samples generated by our approach, comparing it with the original Projected Gradient Descent method and the Carlini & Wagner attack across various parameter configurations. Remarkably, our proposal consistently achieved the best trade-off between mean error and the number of generated adversarial samples, showcasing its superiority in safeguarding industrial systems.

In Industry 4.0, security begins with the workers’ authentication, which can be done individually or in groups. Recently, group authentication is gaining momentum, allowing users to authenticate as group members without the need to specify the particular individual. Continuous authentication and federated learning are promising techniques that might help group authentication by providing privacy, by its own design, and extra security compared to traditional methods based on passwords, tokens, or biometrics. However, these techniques have not previously been combined or evaluated for authenticating workers in Industry 4.0. Thus, this paper proposes a novel continuous group authentication privacy-preserving (CGAPP)platform that is suitable for the industry. The CGAPP platform incorporates statistical data from workers’ smartphones and employs federated learning-based outlier detection for group worker authentication while ensuring the privacy of personal data vectors. A series of experiments were performed to measure the framework’s suitability and address the following research questions: (i) What is the cost of using FL compared to full data access in industrial scenarios? (ii) How robust is federated learning against adversarial attacks, specifically, how much malicious data is required to deceive the model? and (iii) How much noise is required to disrupt the authentication system? The results demonstrate the effectiveness of the CGAPP platform in the industry since it provides factory safety while preserving privacy. This platform achieves an accuracy of 92%, comparable to the 96% obtained by traditional approaches in the literature that do not address privacy concerns. The platform’s robustness is tested against attacks in the second and third experiments, and various countermeasures are evaluated. While the CGAPP platform exhibits certain vulnerabilities to data injection attacks, straightforward countermeasures can alleviate them. Nevertheless, the system’s performance experiences a notable impact in the event of a data perturbation attack, and the countermeasures investigated are ineffective in addressing this issue.

The connectivity and resource-constrained nature of single-board devices open the door to cybersecurity concerns affecting Internet of Things (IoT) scenarios. One of the most important issues is the presence of unauthorized IoT devices that want to impersonate legitimate ones by using identical hardware and software specifications. This situation can provoke sensitive information leakages, data poisoning, or privilege escalation in IoT scenarios. Combining behavioral fingerprinting and Machine/Deep Learning (ML/DL) techniques is a promising approach to identify these malicious spoofing devices by detecting minor performance differences generated by imperfections in manufacturing. However, existing solutions are not suitable for single-board devices since they do not consider their hardware and software limitations, underestimate critical aspects such as fingerprint stability or context changes, and do not explore the potential of ML/DL techniques. To improve it, this work first identifies the essential properties for single-board device identification: uniqueness, stability, diversity, scalability, efficiency, robustness, and security. Then, a novel methodology relies on behavioral fingerprinting to identify identical single-board devices and meet the previous properties. The methodology leverages the different built-in components of the system and ML/DL techniques, comparing the device internal behavior with each other to detect variations that occurred in manufacturing processes. The methodology validation has been performed in a real environment composed of 15 identical Raspberry Pi 4 Model B and 10 Raspberry Pi 3 Model B＋ devices, obtaining a 91.9% average TPR with an XGBoost model and achieving the identification for all devices by setting a 50% threshold in the evaluation process. Finally, a discussion compares the proposed solution with related work, highlighting the fingerprint properties not met, and provides important lessons learned and limitations.

Abstract Traffic accidents are the leading cause of death among young people, a problem that today costs an enormous number of victims. Several technologies have been proposed to prevent accidents, being brain–computer interfaces (BCIs) one of the most promising. In this context, BCIs have been used to detect emotional states, concentration issues, or stressful situations, which could play a fundamental role in the road since they are directly related to the drivers’ decisions. However, there is no extensive literature applying BCIs to detect subjects’ emotions in driving scenarios. In such a context, there are some challenges to be solved, such as (i) the impact of performing a driving task on the emotion detection and (ii) which emotions are more detectable in driving scenarios. To improve these challenges, this work proposes a framework focused on detecting emotions using electroencephalography with machine learning and deep learning algorithms. In addition, a use case has been designed where two scenarios are presented. The first scenario consists in listening to sounds as the primary task to perform, while in the second scenario listening to sound becomes a secondary task, being the primary task using a driving simulator. In this way, it is intended to demonstrate whether BCIs are useful in this driving scenario. The results improve those existing in the literature, achieving 99% accuracy for the detection of two emotions (non-stimuli and angry), 93% for three emotions (non-stimuli, angry and neutral) and 75% for four emotions (non-stimuli, angry, neutral and joy).

This demo presents RITUAL, a platform composed of a novel algorithm and a Web application quantifying the trustworthiness level of supervised Machine and Deep Learning (ML/DL) models according to their fairness, explainability, robustness, and accountability. The algorithm is deployed on a Web application to allow users to quantify and compare the trustworthiness of their ML/DL models. Finally, a scenario with ML/DL models classifying network cyberattacks demonstrates the platform applicability.

Data confidentiality protection is a must for IoT and crowdsensing platforms, and a challenge due to the constrained nature of their sensors. Currently, the combination of device fingerprinting and anomaly detection systems based on Machine and Deep Learning (ML/DL) techniques is one of the most promising approaches to detect zero-day cyberattacks. However, most of existing work is not suitable for resource-constrained devices or does not deal with cyberattacks affecting data confidentiality of spectrum sensors. Thus, this paper proposes a framework that monitors network interface events of sensors, uses unsupervised learning to create fingerprints, and detects anomalies produced by such cyberattacks. The framework validation has been performed in the crowdsensing platform ElectroSense, where a sensor has been infected by a backdoor leaking different sensitive data during an experiment. A set of unsupervised learning algorithms has been evaluated, being Autoencoder the one showing the best balance when detecting normal behavior and data leakages of different sizes and at frequencies, while providing a reduced detection time and sensor resources consumption.

Traditionally, data centers have been the preferred target for ransomware attacks. However, the increasing number of IoT (Internet-of-Things) devices managing valuable data is attracting the attention of cybercriminals and ransomware towards resource-constrained devices. So far, literature has demonstrated the suitability of monitoring the behavior of devices to detect some malware infections. However, most of these existing solutions have been designed and validated in Windows-based systems without computational restrictions.Thus, this work presents a lightweight policy-based framework that uses behavioral fingerprinting to detect anomalies and classify ransomware affecting resource-constrained and Linux-based sensors. The framework detection capabilities have been validated in a resource-constrained spectrum sensor belonging to ElectroSense, a real crowdsensing platform. In particular, three policies, created as a proof-of-concept, resulted in promising findings in terms of detection performance and time, when identifying anomalies by classifying two recent ransomware samples affecting a Raspberry Pi acting as sensor.

Billions of IoT devices lacking proper security mechanisms have been manufactured and deployed for the last years, and more will come with the development of Beyond 5G technologies. Their vulnerability to malware has motivated the need for efficient techniques to detect infected IoT devices inside networks. With data privacy and integrity becoming a major concern in recent years, increasing with the arrival of 5G and Beyond networks, new technologies such as federated learning and blockchain emerged. They allow training machine learning models with decentralized data while preserving its privacy by design. This work investigates the possibilities enabled by federated learning concerning IoT malware detection and studies security issues inherent to this new learning paradigm. In this context, a framework that uses federated learning to detect malware affecting IoT devices is presented. N-BaIoT, a dataset modeling network traffic of several real IoT devices while affected by malware, has been used to evaluate the proposed framework. Both supervised and unsupervised federated models (multi-layer perceptron and autoencoder) able to detect malware affecting seen and unseen IoT devices of N-BaIoT have been trained and evaluated. Furthermore, their performance has been compared to two traditional approaches. The first one lets each participant locally train a model using only its own data, while the second consists of making the participants share their data with a central entity in charge of training a global model. This comparison has shown that the use of more diverse and large data, as done in the federated and centralized methods, has a considerable positive impact on the model performance. Besides, the federated models, while preserving the participant’s privacy, show similar results as the centralized ones. As an additional contribution and to measure the robustness of the federated approach, an adversarial setup with several malicious participants poisoning the federated model has been considered. The baseline model aggregation averaging step used in most federated learning algorithms appears highly vulnerable to different attacks, even with a single adversary. The performance of other model aggregation functions acting as countermeasures is thus evaluated under the same attack scenarios. These functions provide a significant improvement against malicious participants, but more efforts are still needed to make federated approaches robust.

Invasive Brain-Computer Interfaces (BCIs) are extensively used in medical application scenarios to record, stimulate, or inhibit neural activity with different purposes. An example is the stimulation of some brain areas to reduce the effects generated by Parkinson’s disease. Despite the advances in recent years, cybersecurity on BCIs is an open challenge since attackers can exploit the vulnerabilities of invasive BCIs to induce malicious stimulation or treatment disruption, affecting neuronal activity. In this work, we design and implement a novel neuronal cyberattack called Neuronal Jamming (JAM), which prevents neurons from producing spikes. To implement and measure the JAM impact, and due to the lack of realistic neuronal topologies in mammalians, we have defined a use case using a Convolutional Neural Network (CNN) trained to allow a simulated mouse to exit a particular maze. The resulting model has been translated to a biological neural topology, simulating a portion of a mouse’s visual cortex. The impact of JAM on both biological and artificial networks is measured, analyzing how the attacks can both disrupt the spontaneous neural signaling and the mouse’s capacity to exit the maze. Besides, another contribution of the work focuses on comparing the impacts of both JAM and FLO (an existing neural cyberattack), demonstrating that JAM generates a higher impact in terms of neuronal spike rate. As a final contribution, we discuss whether and how JAM and FLO attacks could induce the effects of neurodegenerative diseases if the implanted BCI had a comprehensive electrode coverage of the targeted brain regions.

As recently reported by the World Health Organization (WHO), the high use of intelligent devices such as smartphones, multimedia systems, or billboards causes an increase in distraction and, consequently, fatal accidents while driving. The use of EEG-based Brain–Computer Interfaces (BCIs) has been proposed as a promising way to detect distractions. However, existing solutions are not well suited for driving scenarios. They do not consider complementary data sources, such as contextual data, nor guarantee realistic scenarios with real-time communications between components. This work proposes an automatic framework for detecting distractions using BCIs and a realistic driving simulator. The framework employs different supervised Machine Learning (ML)-based models on classifying the different types of distractions using Electroencephalography (EEG) and contextual driving data collected by car sensors, such as line crossings or objects detection. This framework has been evaluated using a driving scenario without distractions and a similar one where visual and cognitive distractions are generated for ten subjects. The proposed framework achieved 83.9% F1-score with a binary model and 73% with a multiclass model using EEG, improving 7% in binary classification and 8% in multi-class classification by incorporating contextual driving into the training dataset. Finally, the results were confirmed by a neurophysiological study, which revealed significantly higher voltage in selective attention and multitasking.

Continuous authentication (CA) is a promising approach to authenticate workers and avoid security breaches in the industry, especially in Industry 4.0, where most interaction between workers and devices takes place. However, introducing CA in industries raises the following unsolved questions regarding machine learning (ML) models: its precision and performance; its robustness; and the issue about if or when to retrain the models. To answer these questions, this article explores these issues with a proposed supervised versus nonsupervised ML-based CA system that uses sensors, applications statistics, or speaker data collected by the operator’s devices. Experiments show supervised models with equal error rates of 7.28% using sensors data, 9.29% with statistics, and 0.31% with voice, a significant improvement of 71.97, 62.14, and 97.08%, respectively, over unsupervised models. Voice is the most robust dimension when adding new workers, with less than 2% of false acceptance rate even if workforce size is doubled.

Summary Cyber attacks are becoming a very common trend in existing networks, expecting to be even more acute in future 5G networks due to greater number of connected devices, higher mobile data volume, low latency, etc. Security mechanisms to tackle cyber threats should be updated when users, possibly carrying devices with some kind of malware, move in highly dynamic mobile networks in order to continue providing the same detection and mitigation capabilities along user's path. This paper presents BotsOnFIRE, an experiment of the EU H2020 SoftFIRE project for the detection and mitigation of botnets in the SoftFIRE federated testbed for 5G within FIRE+, by combining Software‐Defined Networking (SDN) and Network Functions Virtualization (NFV) technologies. Bots' mobility is considered to trigger reconfiguration of security‐related SDN/NFV applications, when needed, so as to update security capabilities of the SoftFIRE infrastructure. The BotsOnFIRE experiment contributes to the wider 5G objective of more secure and resilient networks and services, where botnets are actually one of the most powerful cyber threats capable of orchestrating the remote execution of cyber‐attacks. Experiments confirm that BotsOnFIRE is feasible to conduct the expected detection and mitigation procedures in the SoftFIRE federated environment, being evaluated through some Key Performance Indicators.

Abstract The evolution of integrated clinical environments (ICE) and the future generations of mobile networks brings to reality the hospitals of the future and their innovative clinical scenarios. The mobile edge computing paradigm together with network function virtualization techniques and the software-defined networking paradigm enable self-management, adaptability, and security of medical devices and data management processes making up clinical environments. However, the logical centralized approach of the SDN control plane and its protocols introduce new vulnerabilities which affect the security of the network infrastructure and the patients’ safety. The paper at hand proposes an SDN/NFV-based architecture for the mobile edge computing infrastructure to detect and mitigate cybersecurity attacks exploiting SDN vulnerabilities of ICE in real time and on-demand. A motivating example and experiments presented in this paper demonstrate the feasibility of of the proposed architecture in a realistic clinical scenario.

Summary Industrial control systems (ICS) are considered cyber‐physical systems that join both cyber and physical worlds. Due to their tight interaction, where humans and robots co‐work and co‐inhabit in the same workspaces and production lines, cyber‐attacks targeting ICS can alter production processes and even bypass safety procedures. As an example, these cyber‐attacks could interrupt physical industrial processes and cause potential injuries to workers. In this article, we present SafeMan, a unified management framework based on the Edge Computing paradigm that provides high‐performance applications for the detection and mitigation of both cyber‐attacks and safety threats in industrial scenarios. Three use cases show specific threats in manufacturing as well as the SafeMan actions carried out to detect and mitigate them. In order to validate our proposal, a pool of experiments was performed with Electra, an industrial dataset with normal network traffic and different cyber‐attacks by using a given number of Modbus TCP and S7Comm devices. The experiments measured the runtime performance of anomaly detection techniques based on machine learning and deep learning to detect cyber‐attacks in control networks. The experimental results show that Neural Networks report the best performance, being able to examine 217 feature vectors per second over Electra, and therefore demonstrating that it can be used as detection model for SafeMan in real scenarios.

The disruptive vision of Medical Cyber-Physical Systems (MCPS) enables the promising next-generation of eHealth systems that are intended to interoperate efficiently, safely, and securely. Safety-critical interconnected systems that analyze patients' vital signs gathered from medical devices, infer the state of the patient's health, and start treatments issuing information to doctors and medical actuators should improve the patients' safety in a cost-efficient fashion. Despite the benefits provided by the MCPS vision, it also opens the door to critical challenges like the security and privacy, Quality of Service (QoS), and high availability of the devices composed to support the MCPS scenario. The Integrated Clinical Environment (ICE) standard is a significant step toward promoting open coordination of heterogeneous medical devices by considering the previous challenges. However, a lot of effort is still required in order to cover the whole aspects of these challenges and enable the future eHealth. In this context, we identify critical shortcomings of ICE using challenge scenarios regarding security, QoS, and high availability. According to these concerns and following the ICE standard, we propose the novel ICE++ architecture, which is oriented to the Mobile Edge Computing paradigm and combines SDN and NFV techniques to manage efficiently and automatically the MCPS elements taking into account its security, QoS, and high availability. Finally, we perform experiments that demonstrate the potential usefulness of our solution regarding the efficient and automatic management of the ICE components.

Users' privacy is a critical challenge for any information management system. The proliferation of mobile devices has promoted the use of context-aware solutions, making the protection of the users' information an even greater challenge. Addressing this requires a given mechanism that allows users to manage and control their personal information. In this sense, this paper proposes a privacy-preserving and context-aware system named MASTERY (Multicontext-Aware System That prEserves the useRs' privacY) that manages the privacy of the users' information in intra- and inter-context scenarios. MASTERY is a trusted third party that suggests to users a pool of privacy policies (profiles) aware to the context in which they are located, who can modify them according to their interests. These policies protect the privacy of the users' information being accessed from others without their consent. The information about users and contexts is managed by using semantic web techniques. This provides a common infrastructure that makes possible to represent, process, and share information between independent systems easily.